SPARC: an efficient way to combine reinforcement learning and supervised autonomy

Senft, Emmanuel and Lemaignan, Severin and Baxter, Paul E. and Belpaeme, Tony and UNSPECIFIED and UNSPECIFIED (2016) SPARC: an efficient way to combine reinforcement learning and supervised autonomy. In: Future of Interactive Learning Machines Workshop at NIPS'16, 5 - 10 Dec 2016, Barcelona, Spain.

Full content URL: http://www.filmnips.com/wp-content/uploads/2016/11...

Documents
16 NIPS Emmanuel.pdf
[img]
[Download]
[img]
Preview
PDF
16 NIPS Emmanuel.pdf - Whole Document

186kB
Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive

Abstract

Shortcomings of reinforcement learning for robot control include the sparsity of the environmental reward function, the high number of trials required before reaching an efficient action policy and the reliance on exploration to gather information about the environment, potentially resulting in undesired actions. These limits can be overcome by adding a human in the loop to provide additional information during the learning phase. In this paper, we propose a novel way to combine human inputs and reinforcement by following the Supervised Progressively Autonomous Robot Competencies (SPARC) approach. We compare this method to the principles of Interactive Reinforcement Learning as proposed by Thomaz and Breazeal. Results from a study involving 40 participants show that using SPARC increases the performance of the learning, reduces the time and number of inputs required for teaching and faces fewer errors during the learning process. These results support the use of SPARC as an efficient method to teach a robot to interact with humans.

Keywords:SPARC, machine learning, human-robot interaction, interactive learning
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
Divisions:College of Science > School of Computer Science
Related URLs:
ID Code:30194
Deposited On:13 Mar 2018 08:24

Repository Staff Only: item control page